The ongoing HIV epidemic increasingly involves non-injection drug use (NIDU). In HIV-infected individuals, particularly vulnerable populations such as low-income minority patients in HIV primary care, reducing NIDU is important to improve overall health and HAART adherence, and to reduce sexual transmission of HIV. A barrier to progress in NIDU reduction for this complex population has been the lack of effective interventions that make limited demands on staff and resources. An intervention designed to overcome this barrier combines Motivational Interviewing (MI) with automated telephone technology (HealthCall). The intervention begins with a counselor administering a brief MI session targeting NIDU. Then, with little additional staff time, HealthCall extends the intervention through patient use of Interactive Voice Response (IVR) technology, which provides behavioral self-monitoring of NIDU and other target behaviors. Patients make brief daily calls to the IVR for 60 days. At 30 and 60 days, MI counselors meet with patients and give them brief personalized feedback, discussing a graph on their drug use generated from the IVR data. Thus, HealthCall serves as a clinician-extender of the brief MI. A successful randomized pilot study of MI+HealthCall among low-SES, minority patients with active non- injection drug use in a large, metropolitan, HIV primary care clinic demonstrated acceptability and feasibility of MI+HealthCall, and, compared to MI-only, suggested efficacy of MI+HealthCall in reducing drug use and improving HAART adherence. A 3-arm randomized controlled trial (RCT) of MI+HealthCall is proposed in the same setting and patient population. In this RCT, 390 patients will be randomized to (1) MI+HealthCall, (2) MI-only or (3) educational control. Evaluations will be done at baseline, 30 and 60 days, and at 3, 6 and 12 months. Main outcomes are reduction in use of primary drug (non-injection cocaine, heroin or methamphetamine), improved HAART adherence, and reduced sexual risk behavior. Mediators (motivation to change and self-efficacy) and moderators (e.g., depression) will be explored. Achieving study aims could help reduce NIDU, improve HAART adherence and reduce sexual risk behaviors in HIV primary care patients. More broadly, a novel, technologically-enhanced intervention would be added to the treatment literature that could change concepts about the ratio of required clinician time to the duration of patient participation in an intervention. This proposal is responsive to NIDA PA-09-236 HIV/AIDS, Drug Use, and Vulnerable Populations in the US.